© Thinkstock

The number of foreign-born people in the United States now constitutes 12 percent of the population--the highest share since about 1930. The rise in the number of recently arrived residents--nearly half of the immigrants in the United States have arrived since 1990--has raised broad questions about the potential effects of immigration on labor markets and economic performance in general. Immigration increases the pressures for federal, state, and local government spending. Immigrants also contribute to the economy and pay taxes. A major question is whether immigration has the potential to lessen the strain on the federal budget as the baby-boom generation retires.

This paper is the first of several reports by the Congressional Budget Office (CBO) intended to present the facts and research on immigration to help inform the agency's projections of the federal budget and the economy. The paper focuses on the characteristics of immigrants, including where they come from, where they reside, and how those patterns have changed over time, as well as immigrants' level of education, the industries and occupations in which they work, and their earnings. In keeping with CBO's mandate to provide objective, nonpartisan analysis, this paper makes no recommendations.

Douglas Hamilton is coordinating CBO's series of reports on immigration. David Brauer wrote this paper with research assistance from Tumi Coker, Adam Gordon, and Amrita Palriwala. Carol Frost provided programming advice. Nabeel Alsalam, Selena Caldera, Paul Cullinan, Bob Dennis, Meena Fernandes, Teri Gullo, Arlene Holen, Melissa Merrell, Noah Meyerson, John Peterson, Elizabeth Robinson, J.C. Rodriguez, Kathy Ruffing, and Ralph Smith provided comments on early drafts of the paper, as did Joseph M. Costanzo of the Census Bureau. (The assistance of external reviewers implies no responsibility for the final product, which rests solely with CBO.)

Janey Cohen edited the paper, and John Skeen proofread it. Maureen Costantino prepared the paper for publication and designed the cover, Lenny Skutnik printed the initial copies, and Annette Kalicki produced the electronic version for CBO's Web site.

Douglas Holtz-Eakin
Director
November 2004

Summary

The Census Bureau estimated that the number of foreign-born people living in the United States topped 33 million and accounted for nearly 12 percent of the population in 2003--its highest share since 1930. Half of those people have arrived in the United States since 1990, and the foreign-born population is now growing at a rate of about 1 million per year. Between the 1990 and 2000 censuses, the foreign-born population grew by more than 11 million, accounting for about 35 percent of total population growth. Estimates of the portion of immigrants who are unauthorized range from 7 million to 10 million, although in the absence of reliable data, those estimates are subject to considerable uncertainty.

More than half of the total foreign-born population originated in Latin America, with Mexico by far the single largest source. The number of immigrants from Asia has also grown rapidly since the 1960s. In contrast, in the 19th and early 20th centuries, the vast majority of immigrants came from Europe.

The foreign-born population is quite concentrated geographically, more so than people born in the United States. More than two-thirds of the foreign-born population reside in one of six states (California, Texas, New York, Florida, Illinois, and New Jersey), and the majority of newly admitted legal immigrants continue to settle in those states. However, that concentration has diminished somewhat since 1990, and new major destinations include Arizona, Nevada, North Carolina, Georgia, and Colorado. The foreign-born population also tends to be more concentrated in urban areas--especially central cities--than the native-born is, but as is the case generally in the United States, more than half of the foreign-born population now live in suburban areas.

Because children are underrepresented in the foreign-born population, the median age of the foreign-born population is older than that of the native-born population. However, foreign-born adults as a group are younger than native-born adults. That difference is greatest for those from Latin America and exists to a lesser extent for those from Asia.

The skill distribution among the foreign-born population is bifurcated, with both a much larger fraction than natives lacking a high school diploma, yet a slightly higher than average percentage having an advanced degree. Immigrants from Asia more frequently possess a bachelor's or higher degree, while less than half of those from Latin America have completed high school.

Nearly half of the growth in the labor force since 1995 is attributable to immigrants. Foreign-born men are active participants in the labor market, with those from Latin America more heavily represented in service and blue-collar occupations and those from Europe and Asia in professional occupations. However, foreign-born women are less likely to participate in the labor force than are women born in the United States.

The earnings of foreign-born workers, particularly non-U.S. citizens, are on average well below those of natives. Immigrants from Latin America are predominantly at the low end of the earnings distribution. That circumstance translates into below-average incomes and a higher incidence of poverty in households with foreign-born heads. However, the earnings gap tends to diminish as immigrants spend more time in the United States.
 

How Large Is the Foreign-Born Population?

Data from the Census Bureau's Current Population Survey (CPS) indicate that the foreign-born population living in the United States stood at about 33 million in March 2003 (see Box 1 for a more complete definition of key terms).⁽¹⁾ Of that total, about 38 percent had become U.S. citizens.
 

Box 1. Key Definitions The foreign-born population, as defined by the Census Bureau, refers to all residents of the United States who were not U.S. citizens at birth, regardless of their current legal or citizenship status. Natives are those who were born in one of the following areas--the United States, Puerto Rico, Guam, American Samoa, the U.S. Virgin Islands, or the Northern Mariana Islands--or were born abroad to at least one parent who was a U.S. citizen. All residents are either natives or foreign born, but not both. Immigrants are defined by the Office of Immigration Statistics (OIS) as persons legally admitted to the United States as permanent residents. Refugees and asylees, as defined by OIS, are people admitted to the United States because they are unable or unwilling to return to their country of nationality because of persecution or a well-founded fear of persecution. Refugees apply for admission at an overseas facility and enter the United States only after their application is granted; asylees apply for admission when already in the United States or at a point of entry. Legal temporary residents are foreign citizens authorized to enter and reside temporarily in the United States for a specific purpose. Unauthorized immigrants are foreign citizens illegally residing in the United States. They include both those who entered without inspection and those who violated the terms of a temporary admission without having gained either permanent resident status or temporary protection from removal. (Also referred to as illegal or undocumented immigrants.) Nonimmigrants, as defined by OIS, are foreign citizens admitted to the United States for a specified purpose and a temporary period, including both legal temporary residents and visitors. The residual foreign-born population consists of foreign-born residents who are neither naturalized citizens nor legal permanent residents.

The total foreign-born population stood at 2 million in the 1850 census, then grew rapidly over the next 60 years (see Figure 1). Subsequently, the rate of growth slowed substantially before the population reached 14 million in 1930. The foreign-born population subsequently declined, reaching a 20th century low of less than 10 million in 1970. Since then, however, the foreign-born population has more than tripled, with an average annual growth rate of about 4 percent. As a share of the total population, the number of foreign-born people rose from 10 percent to 13 percent between 1850 and 1860, then remained in the 13 percent to 15 percent range through 1920. That share then plunged to just under 5 percent in 1970 but has since rebounded steadily and substantially. By 2002, it stood at nearly 12 percent--the highest share since about 1930.

Figure 1.

Source: Congressional Budget Office using data from Campbell J. Gibson and Emily Lennon, Historical Census Statistics on the Foreign-Born Population of the United States: 1850-1990, Department of Commerce, Bureau of the Census, Population Division, Working Paper No. 29 (February 1999); and Nolan Malone and others, The Foreign-Born Population: 2000, Census 2000 Brief (Department of Commerce, Bureau of the Census, December 2003).

Those census estimates are measures of the total number of immigrants in the population. Changes in that number represent net inflows--that is, the number of people newly arriving minus the number of residents who emigrate from the United States. During the 1990s, the total foreign-born population rose by 11 million, or slightly over 1.1 million annually, with net inflows apparently accelerating toward the end of the decade. As a consequence, in its latest population estimates, the Census Bureau assumed that net migration into the United States was about 1.3 million annually between 2000 and 2003. The Census Bureau arrived at this figure by separately estimating the number of newly arriving legal immigrants, refugees, and asylees; the number of foreign-born U.S. citizens and legal permanent residents who emigrated; and net flows of nonimmigrants (which include unauthorized residents as well as legal temporary residents and those with quasi-legal status).⁽²⁾

Legal Immigrants, Refugees, and Asylees

In considering flows of legal immigrants, one must distinguish between people who physically arrive in the United States and people who are awarded legal permanent resident status and thus are legally admitted as immigrants. U.S. Citizenship and Immigration Services (CIS) admitted about 700,000 people as immigrants in fiscal year 2003, down sharply from the more than 1 million admitted in both 2001 and 2002 (see Table 1).⁽³⁾ However, only about half of those admitted in 2003, and slightly more than a third of those admitted in 2002, were new arrivals. Indeed, last year's decline in the total number of immigrants admitted to the United States primarily stemmed from a reduction in the number of people adjusting to immigrant status after arrival rather than in the number of new arrivals. Many of those people became immigrants after entering the United States legally as nonimmigrants--for instance, as visitors, temporary workers, students, workers affected by intracompany transfers, fiancés or fiancées of U.S. citizens, refugees, and asylees.

Table 1.

Most immigrants admitted to the United States are sponsored by their families. In 2003, more than two-thirds were admitted as family-sponsored immigrants, with the majority as immediate relatives (spouses, children, or parents) of U.S. citizens.⁽⁴⁾ The family preference category consists of adult sons and daughters of U.S. citizens, as well as their spouses and children; spouses and children of permanent residents; and siblings of U.S. citizens. By contrast, only 12 percent were admitted under specific employment-based preferences, and just 6 percent were refugees and asylees who adjusted to immigrant status.

According to CIS's records, 9.1 million people were admitted as legal immigrants during the 1990s--the largest number of immigrants admitted in any decade since records were kept beginning in 1820, and significantly more than the 7.3 million admitted during the previous decade.⁽⁵⁾ The level of immigration rose from negligible levels in the 1820s until it exceeded 5 million during the 1880s (see Figure 2). Then, after falling back during the 1890s, the immigration level peaked at 8.8 million between 1901 and 1910. From there it dropped off dramatically, reaching a low of just over 500,000 in the 1930s before rebounding at first gradually and then more rapidly in the late 20th century until it surpassed its previous high during the 1990s. When placed in the context of a large and growing population, the impact of immigration during the 1990s was considerably smaller than in earlier periods of history. The number of immigrants during the 1990s represented 3.7 percent of the 1990 population, up from 3.2 percent during the 1980s but well below the 11.6 percent peak attained between 1900 and 1910. The number of immigrants also exceeded 10 percent of the population (as measured at the start of the decade) in both the 1850s and 1880s.

Figure 2.

Source: Department of Homeland Security, Office of Immigration Statistics, 2003 Yearbook of Immigration Statistics (September 2004).

Note: Arrivals by land were not completely enumerated until 1908.

Aside from those admitted as legal immigrants, about 28,000 people were newly admitted as refugees in fiscal year 2003, and another 15,000 were granted asylum. Refugees normally apply for admission at an overseas facility (such as an embassy), and travel to the United States only after their application is granted. By contrast, requests for asylum can come only from people already in the United States (whether legally present in the United States or otherwise) or who present themselves to authorities at a border crossing or other point of entry.⁽⁶⁾ The number of refugees was slightly above that of a year earlier but much lower than the 69,000 figure for fiscal year 2001. That decline was attributed to heightened safety concerns and enhanced security procedures following the terrorist attacks on September 11, 2001, which affected both the processing of applications at some overseas locations and the travel of those whose applications had already been approved. The number of people granted asylum was little changed between 2001 and 2002 but declined significantly in 2003.

Visitors and Temporary Residents

Aside from immigrants, refugees, and asylees, an estimated 181 million people were admitted to the United States as nonimmigrants in 2003. The vast majority were short-term visitors from Canada and Mexico who were not required to fill out arrival forms. Excluding those visitors, nearly 28 million people were recorded as having entered the United States as nonimmigrants in 2003. Of those, the vast majority were temporary visitors--mostly for pleasure (see Table 2). Those figures were essentially unchanged from 2002 but about 5 million below their 2001 levels. Another 550,000 were admitted as transit aliens--aliens in immediate and continuous transit through the United States. Roughly 3 million people were admitted as temporary residents, including about 950,000 admitted either as temporary workers or as workers affected by intracompany transfers (plus their nearly 300,000 family members); 625,000 admitted as students; and 320,000 as exchange visitors.

Table 2.

Emigration

Because information on departures is not systematically recorded, estimates of emigration must necessarily be derived from incomplete source data. In fact, as noted earlier, the Census Bureau attempts to estimate emigration only by legal permanent residents; for temporary residents and other nonimmigrants, it estimates net flows. The best estimates suggest that one-fourth to one-third of legal immigrants eventually leave the United States, with most emigration occurring within several years of admission. Census Bureau researchers estimated that an average of about 190,000 foreign-born residents per year emigrated during the 1980s, and about 220,000 per year emigrated during the 1990s.⁽⁷⁾ (In addition, the Census Bureau assumes that about 50,000 people born in the United States emigrate annually to other countries.)

Unauthorized Immigrants

Deriving estimates of the number of unauthorized, or illegal, immigrants is difficult because the government lacks administrative records of their arrival and departure, and because they tend to be undercounted in the census and other surveys of the population. Unauthorized immigrants generally fall into one of two categories: those who entered the United States illegally and without inspection and those who were admitted legally as visitors or temporary residents but overstayed their visa.

The unauthorized resident population was recently estimated by the Immigration and Naturalization Service (INS)--whose immigrant services operations are now performed by CIS--at 7.0 million in January 2000, up from 3.5 million in 1990. Alternatively, researchers at the Urban Institute estimated an undocumented population of 9.3 million in March 2002.⁽⁸⁾ A third estimate, by the Census Bureau, put the "nonimmigrant" population at 8.7 million in 2000. However, in addition to unauthorized immigrants, that estimate includes refugees and asylees who have not yet adjusted to legal permanent resident status, and others who are awaiting action on applications to become immigrants or who have been permitted to stay in the United States under court order.

INS's estimate implies that during the 1990s, the number of illegal immigrants in the United States rose by 3.5 million. The gross inflow over the decade was estimated at a much larger 7.1 million, but that number was reduced by voluntary emigration, subsequent adjustments to legal status, and, to a lesser extent, deportations and deaths (for further discussion of estimates of the unauthorized population, see Box 2).
 

Box 2. Estimating the Population of Unauthorized Immigrants In the absence of records on the number of unauthorized immigrants, their numbers must be estimated by indirect means. Thus, those estimates are subject to a considerable degree of uncertainty. The typical method is to start with a base year estimate of the unauthorized immigrant population, subtract the number who subsequently left that population, add the number of newly arrived foreign-born residents, then subtract those within that group who are known to be in the United States legally. But that approach has problems, including uncertainty over the rate at which that population is undercounted in the census, the need to make assumptions about emigration and mortality, and the treatment of immigrants in the United States legally but whose status is ambiguous. The Immigration and Naturalization Service (INS) estimated that 7.0 million unauthorized residents lived in the United States in January 2000.(1) To obtain that figure, INS developed an estimate of the number of unauthorized residents who entered the United States during the 1990s and separately estimated the number who entered before 1990 and still lived here illegally in 2000. Its estimate of the number who entered during the 1990s was derived by subtracting estimates of the legally resident foreign-born population who entered between 1990 and 1999 from the total foreign-born population who arrived during the 1990s, with an adjustment for an estimated undercount in the census. For unauthorized residents who entered before 1990, the estimate was based on the population believed to have resided illegally in January 1990, reduced by the number who left the unauthorized population during the 1990s. INS's latest estimate for 2000 begins with about 3.5 million unauthorized immigrants present in the United States in January 1990. Of those, 2.0 million left the unauthorized population (through adjustment to legal status, emigration, or death), yielding a total of 1.5 million still illegally residing in the United States. Separately, the 2000 census counted 12.6 million foreign-born residents who reported having entered between 1990 and 1999. Adjusting for the census undercount and suspected misreporting of year of entry yields an estimated foreign-born population of nearly 13.5 million who entered in the 1990s.(2) But according to INS, 5.8 million of those people were admitted as legal permanent residents, 0.9 million as refugees, 1.5 million as legal nonimmigrants, and 0.5 million in other categories, including asylees and special agricultural workers. Another 200,000 were unauthorized but applied for legal permanent residency; in those estimates they are treated as legal residents. Adjusting for deaths, emigration, and misclassification of new arrivals results in an estimate of 8.0 million legally resident foreign-born people who arrived between 1990 and 1999. That leaves 5.5 million unauthorized new arrivals, which, when added to the 1.5 million who entered before 1990, yields an estimate of 7.0 million. The approach underlying the latest estimates differs from INS's previous estimates, which were based primarily on estimates of unauthorized residents counted in the 1980 census; administrative records of adjustments to lawful status under the Immigration Reform and Control Act (IRCA) of 1986; and estimates of nonimmigrants overstaying from the INS's Nonimmigrant Information System, which was designed to record arrivals and departures of visitors and temporary residents. However, the collection and processing of departure forms were incomplete, making it hard to distinguish overstaying from a failure to collect departure forms and match them to arrival forms. The last estimate using the old methodology showed about 5.0 million unauthorized immigrants residing in the United States in October 1996; the latest estimates suggest 5.8 million in that month. An alternative calculation by INS sheds light on the gross flows of unauthorized immigrants. In the 1990s, about 7.1 million people entered the unauthorized population but 3.6 million left via emigration (1.5 million), death (200,000), removal by INS (400,000), or adjustment to legal status (1.5 million).(3) The estimated number of entrants rose from 556,000 in 1992 to 772,000 in 1995, retreated over the next two years, then surged to 968,000 in 1999. Rather than estimating the number of unauthorized residents, the Census Bureau estimated the number of "nonimmigrants"--all foreign-born residents except naturalized U.S. citizens and legal permanent residents.(4) Thus, in addition to unauthorized residents, its estimate includes refugees and asylees who have not adjusted to legal permanent resident status, plus others with pending applications for permanent resident status. The bureau's approach was similar to INS's latest approach in that it estimated the nonimmigrant population as a residual, subtracting the number of naturalized citizens and legal permanent residents from the total foreign-born population. It first estimated a residual foreign-born population in 1990 by taking an earlier estimate of 2.1 million in 1980, which dropped to 1.8 million after adjusting for deaths and emigration. It then added the number of foreign-born residents counted in the 1990 census, subtracting those legally admitted between 1980 and 1990 as well as those legalized under IRCA and temporary residents admitted between 1987 and 1990. Adjusting for misclassification, deaths, and emigration yielded a 1990 estimate of 3.8 million. For 2000, the bureau carried forward its 1990 estimate reduced by emigration, deaths, and adjustments to legal status, then followed the same methodology as in the 1990 estimate to determine the residual population entering during the 1990s. The result was an overall estimate of 8.7 million. However, bureau analysts noted a number of limitations to their estimate. First, it does not adjust for the undercount; a 15 percent undercount adjustment would yield estimates of 4.4 million in 1990 and 10.2 million in 2000. A 20 percent undercount would increase those totals to 4.7 million and 10.9 million, respectively. Also, the bureau's estimates were based on incomplete data from the 2000 census: in particular, the number of nonimmigrant legal residents counted in 2000 appears to have been too low, implying an overstatement of the residual population. Further, the Census Bureau's estimate of the residual population in 1980--the starting point for its analysis--is subject to uncertainty. Other limitations that apply equally to INS's estimates involve uncertainty about emigration and mortality rates among the unauthorized or residual population. And the inclusion of quasi-legal migrants awaiting action on their legal migration requests makes direct comparisons with INS's data on unauthorized residents problematic. A third estimate, by researchers at the Urban Institute, shows 9.3 million "undocumented immigrants" in March 2002.(5) That estimate assumes that the total foreign-born population was undercounted by about 2 million in the March 2002 Current Population Survey. The authors note, however, that their estimate includes some people--possibly up to 10 percent of their total estimate--who were in fact authorized to live and work in the United States. They also note that this population includes an estimated 1.6 million children under 18, and that another 3 million U.S.-born children have parents who are undocumented immigrants.(6) 1.  Immigration and Naturalization Service, Office of Policy and Planning, Estimates of the Unauthorized Immigrant Population Residing in the United States: 1990 to 2000 (January 2003), available at http://uscis.gov/graphics/shared/aboutus/statistics/Ill_Report_1211.pdf. 2.  INS assumes an overall undercount of 2.5 percent for legal permanent residents, 10 percent for unauthorized residents, and 10 percent for all other categories of non-U.S. citizens. 3.  Estimates of the number of people leaving the unauthorized resident population include both those who were present in January 1990 and those who entered subsequently. Some people may be counted as entering and/or leaving the unauthorized population more than once during the decade. 4.  Joseph M. Costanzo and others, Evaluating Components of International Migration: The Residual Foreign Born, Working Paper No. 61 (Bureau of the Census, Population Division, June 2002). 5.  Jeffrey S. Passel, Randolph Capps, and Michael E. Fix, "Undocumented Immigrants: Facts and Figures" (Washington, D.C.: Urban Institute, January 2004). 6.  The 3 million U.S.-born children of undocumented immigrants are not included in the 9.3 million estimate of the undocumented population.

 

Where Does the Immigrant Population Come From?

Data from the March 2003 CPS indicate that 53 percent of foreign-born residents were born in Latin America, 25 percent in Asia, and 14 percent in Europe (see Table 3). (As will become evident later in this paper, the demographic characteristics and indicators of skill vary significantly by immigrants' region of origin.) Many of the foreign-born residents have arrived fairly recently: about half have entered the United States since 1990.

Table 3.

In general, immigrants from Europe were the most likely, and those from Latin America the least likely, to have taken U.S. citizenship. Given that most permanent residents must wait at least five years for naturalization, it is not surprising that this pattern is largely linked to period of arrival, with Europeans as a group having arrived much earlier than those from other regions. If the period of arrival is taken into account, immigrants from Asia are slightly more likely than Europeans, and substantially more likely than those of Latin American origin, to have attained U.S. citizenship. Overall, about 38 percent of the foreign-born population has been naturalized.

The 2000 census provides more detail on the foreign-born population's country of origin (see Table 4). Mexico was by far the largest single source, with the 9.2 million Mexican-born residents accounting for nearly 30 percent of the foreign-born population. The next nine largest source countries together accounted for 29 percent of the foreign-born population, with China ranking a distant second as the birthplace of 1.5 million residents, followed by the Philippines with 1.4 million and India and Vietnam with 1 million each. Estimates of the legal permanent resident population recently released by the Office of Immigration Statistics present a mostly similar picture.⁽⁹⁾ Of the estimated 11.4 million total population of legal permanent residents, about 3.1 million were from Mexico. The Philippines was a distant second with just over 500,000, followed closely by India, the People's Republic of China, the Dominican Republic, and Vietnam.

Table 4.

The pattern of immigration to the United States has shifted dramatically over the past nearly two centuries. Throughout the 19th century and into the early part of the 20th century, the vast majority of newly admitted immigrants came from Europe (see Table 5 and Figure 3). Until 1890, the most important single-country sources of immigrants were Ireland, Germany, and the United Kingdom. Over the next three decades, the surge in the total number of immigrants could be attributed largely to Italy and Eastern Europe.⁽¹⁰⁾ However, the number of immigrants from Europe declined dramatically after limits on the total number of immigrants were imposed under the Quota Law of 1921. Over the next several decades, overall immigration remained at very low levels and was dominated by Germany and Canada. The rapid rise in immigration starting in the late 1960s was largely attributable to the Western Hemisphere--especially Mexico--and Asia.

Figure 3.

Source: Department of Homeland Security, Office of Immigration Statistics, 2003 Yearbook of Immigration Statistics (September 2004).

Notes: Arrivals by land were not completely enumerated until 1908.

The Americas comprise Latin America and Canada.

Where Does the Foreign-Born Population Live?

The foreign-born population is quite concentrated geographically, though less so than in 1990. Some 9 million of the 31 million foreign-born people counted in the 2000 census lived in California; more than half of that population were in three states (California, New York, and Texas); and more than two-thirds were in six states (those three states plus Florida, Illinois, and New Jersey) (see Table 6). Recent immigrants are continuing to settle in those states; about 63 percent of those newly admitted in 2003 listed as their intended residence one of those states. Moreover, in 2000, INS estimated that nearly half of the unauthorized population was in either California or Texas.

Table 6.

The concentration of foreign-born people in those states is partly attributable to each state's size, but also reflects an above-average foreign-born share of the population within those states. That is especially true of California, where 26.2 percent of the population in 2000 was foreign born, the highest percentage of any state and more than double the national figure of 11.1 percent (see Table 7). New York ranked second in terms of both the total foreign-born population and the foreign-born share of the overall population. However, several smaller states, including Hawaii, Nevada, and Arizona, also had immigrant population shares significantly above the national average. On the other hand, the foreign-born population represented 2 percent of the population or less in seven states.

Table 7.

Over the past 10 years, migration patterns have shifted somewhat, and some new states have emerged as important destinations for the immigrant population. Between 1990 and 2000, the foreign-born population more than tripled in three states (North Carolina, Georgia, and Nevada) and more than doubled in 16 others--none of which were among the six states where two-thirds of the foreign-born population is concentrated (see Table 8). Measured in percentage terms, the growth in the foreign-born population during the 1990s was especially pronounced in the Southwest and Southeast. Although the foreign-born share of the population in 2000 was still below the national average in both North Carolina and Georgia, it rose dramatically between 1990 and 2000--from 1.7 percent to 5.3 percent in North Carolina, and from 2.7 percent to 7.1 percent in Georgia. And in Nevada, the foreign-born share surged from 8.7 percent--just slightly above the national average--in 1990 to 15.8 percent in 2000.

Table 8.

A comparison between the 1990 and 2000 censuses also reveals significantly less geographic concentration among the most recent arrivals. Thirty-eight percent of the immigrants who arrived between 1985 and 1990 and who were counted in the 1990 census resided in California, but only 22 percent of those who arrived between 1995 and 2000 lived in California in 2000.⁽¹¹⁾ In fact, although California still represented by far the largest single destination state in the late 1990s, it was the only state in which the absolute number of entrants was smaller than it had been a decade earlier. Much of the change was driven by recent immigrants from Mexico, whose overall numbers roughly doubled to about 2.5 million. With California's share of the recent arrivals from Mexico declining from 61 percent to 31 percent between 1990 and 2000, virtually all of the increased flow was dispersed among other states, including both "traditional" destinations (notably Texas and Arizona) and "nontraditional" destinations (notably North Carolina, Georgia, and Colorado).

The composition of the foreign-born population across states and regions varies by region of origin. Mexicans, who in 2000 constituted about 30 percent of the total foreign-born population, accounted for 72 percent of the foreign-born population in New Mexico, 66 percent in Arizona, and 65 percent in Texas, but just 4 percent of that population in the Northeast. In contrast, those from other countries in Latin America accounted for nearly two-thirds of the foreign-born population in Florida, 45 percent in New York, and 38 percent in New Jersey, but 10 percent or less in both the Midwest and West. The European share of the foreign-born population tended to be larger in the Northeast and Midwest than in the South and West. The Asian share of the foreign-born population was somewhat less concentrated than that from Latin America and Europe; its share was largest in the West and Midwest and smallest in the South.

Aside from concentrating in particular states, the foreign-born population tends to gather in urban areas to a significantly greater extent than does the U.S.-born population (see Table 9). In 2003, 44 percent of the foreign-born population lived in central cities, compared with 27 percent of their U.S.-born counterparts. Although immigrants have traditionally settled in central cities, about half now live in suburbs--much like the rest of the population. But only 5 percent of the foreign-born population resided outside of metropolitan areas in 2003--far less than the 20 percent figure for those born in the United States.

Table 9.

According to the 2000 census, 27 percent of the foreign-born population was located in four distinct urban locations--Los Angeles County, California (3.4 million, including 1.5 million in the city of Los Angeles); New York City (2.9 million); Cook County, Illinois (1.1 million, including 629,000 in Chicago); and Miami-Dade County, Florida (1.1 million). Miami-Dade County is particularly notable in that it is the only county in the nation in which more than half of the population in 2000 was foreign born.
 

Demographic Characteristics

On average, the foreign-born population is older than the U.S.-born population: their median age of 38.4 years is more than three years older than natives' median age (see Table 10). However, a closer look at the age distribution reveals that only 9 percent of the foreign-born population is younger than 18 years of age, compared with 28 percent of the native population.⁽¹²⁾ Adjusting for that fact, the adult foreign-born population is in fact younger than the native population. Thirty-six percent of the adult foreign-born population is in the 18-to-34 age bracket, compared with 31 percent of natives; a correspondingly smaller fraction of the foreign-born population is 55 or older. And OIS's statistics indicate that of the immigrants legally admitted in 2003, 37 percent were between the ages of 25 and 39, and 28 percent were under 21, but just 11 percent were 55 or older.

Table 10.

The age distribution varies greatly on the basis of region of origin. The median age of European immigrants is 50 years, much higher than that of any other category even when the number of children is taken into account. At the opposite end of the spectrum, the median age of the Latin American population is slightly lower than that of the U.S.-born population, even with the much smaller fraction of children. Some 43 percent of adults from Latin America are between the ages of 18 and 34, and only 16 percent are 55 or older. The Asian adult population is also somewhat younger than its U.S.-born counterpart.

Foreign-born women have a slightly higher rate of fertility compared with women born in the United States (see Table 11). On average, foreign-born women between the ages of 35 and 44 had given birth to 2.1 children, compared with 1.9 children for U.S.-born women in that age group. That difference partly reflects the fact that there is a higher percentage of foreign-born women within that age group who are or have been married.

Table 11.

Educational Attainment

The skill distribution among the foreign-born population is bifurcated. On the one hand, nearly a third of the foreign-born population ages 25 or older did not complete high school, compared with just 12.5 percent of those born in the United States (see Table 12). On the other hand, the percentage of foreign-born people with at least a bachelor's degree is equal to that of the U.S.-born population, and the percentage with an advanced degree is slightly higher.

Table 12.

The variation in educational attainment across regions of origin is extremely pronounced. Half of Asian immigrants possessed at least a bachelor's degree, including 19 percent with an advanced degree, while the percentage lacking a high school diploma was comparable to that of the U.S.-born population. Europeans were both slightly more likely than natives to lack a high school diploma and to possess a college degree (and much more likely to have attained an advanced degree). At the same time, less than half of those who were born in Latin America had completed high school, and just 12 percent possessed a college degree.
 

Employment and Earnings

Foreign-born men of working age are active participants in the U.S. labor market, but foreign-born women tend to be significantly less attached to the labor market than their native counterparts (see Table 13). Among foreign-born men ages 20 to 64, the employment-to-population ratio in March 2003 was 83 percent, slightly higher than the 80 percent ratio for U.S.-born men in that age group. That reflects a somewhat higher labor force participation rate for foreign-born men than for those born in the United States (partly offset by a slightly higher unemployment rate among the foreign born). Men from Latin America had both a higher participation rate and a higher unemployment rate than did both natives and immigrants from other regions. The picture is quite different for women, as foreign-born women had both a substantially lower participation rate and a higher unemployment rate than native-born women had. Although the differences were again most pronounced for women from Latin America--only 55 percent were employed, compared with 70 percent of women born in the United States--women from Asia also exhibited significantly lower participation and higher unemployment than natives.

Table 13.

Among the employed, the occupational distribution is consistent with the data on educational attainment (see Table 14). Foreign-born workers are more likely than those born in the United States to be in service or blue-collar occupations and less likely to work in managerial, professional, sales, or office occupations. Those differences in occupational distribution are entirely driven by immigrants from Latin America, who, as noted above, have very low levels of formal education. In fact, among Europeans, the occupational distribution is fairly similar to that of the U.S.-born population, with a slightly higher percentage in professional and managerial occupations. And Asians are significantly more likely than either those born in the United States or other foreign-born residents to be in professional occupations.

Table 14.

The distribution of employment by industry also differs somewhat between the foreign-born and U.S.-born populations. Compared with natives, the foreign born are more likely to be employed in construction, manufacturing, and leisure and hospitality, and less commonly employed in financial activities, educational and health services, and public administration. As with occupational distribution, the differences are largely driven by immigrants from Latin America.

One avenue by which new immigrants can succeed in the U.S. economy is by establishing their own businesses. Data from the March 2003 CPS indicate that 6.8 percent of foreign-born workers were self-employed, compared with 7.3 percent of U.S.-born workers. Immigrants from Europe, who have been in the United States the longest, were most likely to be self-employed (9 percent), while those from Latin America were least likely (6 percent).⁽¹³⁾ Results from a 1992 survey of business owners indicated that many of the foreign-born self-employed owned small firms, particularly in retail trade and transportation.

The distribution of skills and occupations is reflected in annual earnings (see Table 15). In 2002, the median annual earnings of foreign-born wage and salary workers who were employed year-round and full time were about $27,000, or 75 percent of median earnings among natives. The gap was larger for men, whose median annual earnings were 71 percent of native men's earnings. Earnings of naturalized citizens were higher than noncitizens' earnings and only slightly below those of natives, while noncitizens' earnings were well below the earnings of natives. As with educational attainment and occupational distribution, differentials in annual earnings between foreign-born and U.S.-born workers were driven entirely by immigrants from Latin America, whose median annual earnings were just 61 percent of the average among natives (56 percent among men).

Table 15.

A more complete examination of annual earnings in 2002 reveals that the entire earnings distribution among the foreign born is skewed toward the low end of the overall distribution (see Table 16). Thirty-one percent of the foreign-born workers who were employed year-round and full time earned less than $20,000, compared with 17 percent of natives. The difference was much less pronounced at the high end of the distribution, as 10 percent of the foreign born and 13 percent of natives earned at least $75,000. The distribution was especially skewed toward the low end for noncitizens and for Latin American natives. By contrast, among naturalized citizens, the overall distribution was fairly similar to that of natives, with a moderately higher concentration in both the upper and lower tails.

Table 16.

Although the relatively low average earnings of immigrants are largely attributable to their observed below-average skill level, the low earnings may also reflect the situation of having to establish oneself in the labor market shortly after arriving in the country. However, as immigrants gain experience, the earnings gap tends to narrow. One study shows that although new immigrants earn significantly less than natives with similar observed skills, on average their relative earnings rise by at least 20 percent over their first 10 years of experience.⁽¹⁴⁾ Other studies indicate smaller but nonetheless significant gains as workers assimilate.⁽¹⁵⁾ Assimilation also appears to continue in the second generation, as children of immigrants tend to obtain more education and to earn more than children of native parents with similar levels of education.⁽¹⁶⁾
 

Income and Poverty Status

The fact that annual earnings are lower for foreign-born workers than for natives translates into lower household income (see Table 17). The median income of families with a foreign-born head was about $43,000--79 percent of the median income of families headed by a native. The median income of families with a head of household born in Latin America was just 64 percent of that of families headed by a native. On the other hand, families headed by immigrants from Asia had higher-than-average income.

Table 17.

The foreign born are more likely to be defined as poor than are natives. Among all ages and both sexes, 17 percent of the foreign born are in households with income below the poverty line in 2002, compared with 12 percent of natives (see Table 18). The differences are most pronounced for children under 18 but are also quite clear for both those in the prime working ages and for the relatively small number of foreign-born people age 65 or older.

Table 18.